Focus of attention in discourse 137(18) a. There was this terrible boy at the circus, who....
b. Now listen to this: a man went into a butcher’s shop and bought a
whole pig...
The distal (unmarked) variant, that, would have the discourse-interactional
value whereby its referent is placed outside the speaker’s subjective sphere
and into a shared, ‘negotiated’ space, in which the speaker’s and ad-
dressee’s stances are in co-alignment. As such, the referent is not
(necessarily) envisaged as representing new information for the addressee,
and thus the somewhat lower degree of insistent urging that s/he attend to it
is predicted (see also Cheshire 1996 and Glover 2000 for similar analyses
of English this and that). The ‘recognitional’ or ‘reminder’ use of that pro-
vides a good illustration (see also the demonstrative term that ‘Love Thy
Neighbour’ thing in example (17) above):
(19) Tony Blair was said to be on the phone last week to his German oppo,
Chancellor Schröder, trying to talk through an upbeat final instalment of
that dismal industrial soap opera called Longbridge.
(The Guardian, 3 May 2000, p. 12)
Finally, the non-demonstrative, inanimate (unaccented) pronoun it sig-
nals not only that its intended referent constitutes shared information, but
also that that information is currently in the addressee’s (and the speaker’s)
ongoing attention focus. Again, this value implies that the addressee need
expend no particular effort at all in locating and retrieving the intended ref-
erent, since it is (assumed to be) ‘right there’, in his/her current
consciousness.
Unlike both the CS account of DEIXIS and the Gundel et al. (1993)
Givenness Hierarchy, which are formulated in terms of particular lexical
items, the four FG Topic types are characterized in terms of purely dis-
course-functional roles: although there are claimed to be certain
prototypical indexical markers which realize such roles (zeros and unac-
cented 3rd person pronouns for GivTops, definite lexically-headed terms
for SubTops, expanded definite or demonstrative terms for ResTops, and
indefinite lexically-headed terms for NewTops), it is also clear that, in
given contexts, other expression types may also be used to realize each of
these functions (for example, the demonstratives, whether as determiners
or pronouns, may in fact realize all of them – except ResTops, where de-
monstrative NPs rather than pronouns are required: see Cornish 1998 for
some discussion, based on an English-language newspaper article).